Jcode::Nihongo(3pm)

✨ Jcode::Nihongo(3pm) ✨

User Contributed Perl Documentation 🎀 okay bestie, this is the manpage for the Jcode Japanese charset handler 💖

✨ 名前 ✨

Jcode - 日本語文字集合ハンドラ 🍓 the Japanese text-handling queen 👑

✨ 概要 ✨

use Jcode;
#
# 古式ゆかしく
Jcode::convert($str, $ocode, $icode, "z");
# もしくはオブジェクト指向で！
print Jcode->new($str)->h2z->tr($from, $to)->utf8;

✨ 説明 ✨

ngl this module is so cute 🥹 Jcode.pm supports BOTH the object-y way and the old-school way, so you can pick whatever matches your vibe 🦋 With the object approach it looks like this, literally so clean:

$iso_2022_jp = Jcode->new($str)->h2z->jis;

The old way is also giving, but honestly the one above is more elegant 💅:

$iso_2022_jp = $str;
&jcode::convert(\$iso_2022_jp, 'jis', &jcode::getcode(\$str), "z");

If objects are new to you, no stress 🫶 Jcode.pm still supports getcode() and convert(), so you're totally covered.

If your Perl is 5.8.1 or newer, Jcode acts as a wrapper around Encode, which is the standard charset handler module since Perl 5.8 ✨ iykyk.

✨ メソッド ✨

Unless I say otherwise, every method here returns a Jcode object 🌸

✨ コンストラクタ ✨

• $j = Jcode->new($str [, $icode]) — makes a Jcode object $j out of $str 🧚 If you don't explicitly set $icode, the input code gets auto-detected, so lowkey magic. For the available charsets, check out getcode further down.

On Perl 5.8.1 and up, $icode can be ANY encoding name that Encode understands 🌈

$j = Jcode->new($european, 'iso-latin1');

When you stringify the object, you get back the EUC-converted string. So instead of "print $j->euc", you can just write "print $j" 💜

• リファレンス渡し — you can pass a reference instead of a scalar value, like this:

Jcode->new(\$str);

This saves a tiny bit of time 💅 The catch is that the value of $str itself gets converted (kinda like $str is "tied" to the jcode object, no cap 🙈).

• $j->set($str [, $icode]) — sets $str as the internal string of $j 🎀 This is clutch when you're reusing the same Jcode object over and over, because you save both object-creation time and memory ✨

# mailboxをSJIS形式に変換
my $jconv = new Jcode;
$/ = 00;
while(<>){
    print $jconv->set(\$_)->mime_decode->sjis;
}

• $j->append($str [, $icode]); — tacks $str onto the end of $j's internal string 🦋
• $j = jcode($str [, $icode]); — the short-and-sweet version of Jcode->new() 🍓 You can do stuff like:

$sjis = jcode($str)->sjis;

✨ 文字列のエンコード ✨

Generally, you get the string converted into the encoding you want with $j->encoded (fill in whatever encoding you're after) 💖

• $sjis = $j->sjis
• $euc = $j->euc
• $jis = $j->jis
• $sjis = $j->sjis
• $ucs2 = $j->ucs2
• $utf8 = $j->utf8 — honestly self-explanatory 🙈 WYCIWYG = What you code is what you get :)
• $iso_2022_jp = $j->iso_2022_jp — this is the same as "$j->h2z->jis" 🧃 Half-width kana are forcibly converted to full-width, no exceptions bestie.

On Perl 5.8.1 and up, you can use any encoding name and alias that Encode supports. For example:

$european = $j->iso_latin1; # メソッド名用に'-'は'_'に置き換えます

Fun fact: Encode::Encoder uses this same little trick 🤓

• $j->fallback($fallback) — On Perl 5.8.1 and up, Jcode holds its internal string as UTF-8 😭 Any character that can't be mapped to the ->encoding gets swapped out for '?'. That's the standard Encode behaviour, so don't panic.

my $unistr = "\x{262f}"; # 陰陽マーク
my $j = jcode($unistr);  # $j->euc の結果は '?' に

You can change this by specifying a fallback, just like in Encode 💅 The values are the same as the ones in Encode. For convenience, "Jcode::FB_PERLQQ", "Jcode::FB_XMLCREF" and "Jcode::FB_HTMLCREF" are aliases for their Encode counterparts ✨

print $j->fallback(Jcode::FB_PERLQQ)->euc;   # '\x{262f}'
print $j->fallback(Jcode::FB_XMLCREF)->euc;  # '☯'
print $j->fallback(Jcode::FB_HTMLCREF)->euc; # '☯'

The default fallback is stored in the global variable $Jcode::FALLBACK, so you can override it by assigning to it:

$Jcode::FALLBACK = Jcode::FB_PERLQQ; # デフォルトのフォールバックスキーマを設定

• [@lines =] $jcode->jfold([$width, $newline_str, $kref]) — folds the lines inside the jcode string every $width characters (default: 72) 🌷 $width counts "half-width" characters, so a full-width character counts as 2 ✨

The newline given by $newline_str (default is "\n") gets added in. It returns all the folded lines as a list 📜

On Perl 5.8.1 and up, passing an array reference of EUC strings as the third argument $kref gives you a simple line-head kinsoku shori (the hanging-punctuation thing called ぶら下がり) 🫧

• $length = $jcode->jlength(); — returns the character count, not the byte length, counting each full-width character as one character 💖

💀 Heads up: this is the first half of the page, so there's more juicy stuff coming next. Stay tuned, bestie 🧚

✨ サブルーチン ✨

($code, [$nmatch]) = getcode($str)

okay so this one is a subroutine 💅 It hands you back the character encoding of $str. The possible codes are:

• ascii — Ascii, with no Japanese codes in it 🌸
• binary — Binary, so it's not a text file 🧱
• euc — EUC-JP 🇯🇵
• sjis — SHIFT_JIS 🍙
• jis — JIS (ISO-2022-JP) 🎀
• ucs2 — UCS2 (Raw Unicode) 🦋
• utf8 — UTF8 💖

If you call it in list context instead of scalar context, you also get back how many characters' worth of codes it found. Same deal as before: $str can be \$str too, bestie ✨

jcode.pl people, listen up 👂 This function is 100% backwards compatible with jcode::getcode(). Well, almost 100% 🙈

• when the return value is a list, the order is reversed 🔄. jcode::getcode() gives you $nmatch first.
• jcode::getcode() returns 'undef' when the number of EUC characters and SJIS characters is equal 💀. Jcode::getcode() returns EUC in that case instead, because Jcode.pm has no middle ground 🫠

Jcode::convert($str, [$ocode, $icode, $opt])

This one converts $str into the character code you name with $ocode 🔁. If you also pass $icode, it just assumes your input is that code and skips the getcode() check entirely ✨ Again, $str can be \$str 💜

jcode.pl people 👀 This function is 100% backwards compatible with jcode::convert()! No "almost" this time 💅🔥

---

✨ バグ ✨

Heads up 🚨 When you're on Perl 5.8.1 or newer, Jcode acts as a wrapper around Encode. That means any bugs in Encode can hit Jcode too. Not a Jcode problem, a whole-chain problem 😭

---

✨ 謝辞 ✨

This package owes a lot to 歌代 和正 <utashiro@iij.ad.jp> 🫶 and their Perl4 library jcode.pl, for the motivation, the design, and a big chunk of the code. Literally the foundation of everything 🏛️

大崎 博基 <ohzaki@iod.ricoh.co.jp> was a huge help with polishing the regular expressions, right from the super early days 💎

makamaka@donzoko.net and their JEncode gave me a major lightbulb moment about how to merge Jcode and Encode 💡 They also donated the Japanese manual, which is so sweet 🥹

And the whole Jcode ML <jcode5@ring.gr.jp> crew 🌈 Honestly this thing would not be finished without you all 💜

---

✨ 参考 ✨

• Encode 🧚
• <http://www.iana.org/assignments/character-sets> 📚

---

✨ 著作権 ✨

Copyright 1999-2005 Dan Kogai <dankogai@dan.co.jp> 👑

This library is free software 🕊️ You can redistribute it and/or modify it under the same terms as Perl itself. Main character energy, legally speaking 💅

---

That's the whole page, bestie 🎀 Pulled from Perl v5.40.1, dated 2025-06-11, for Jcode::Nihongo(3pm). Go forth and convert those encodings, you absolute legend 🦋✨